Project Summary/Abstract Genome-wide association studies (GWAS) have been successful in identifying genetic variants affecting the risk of common diseases. GWAS have identified thousands of associated variants, and in some instances the underlying causal variants have been fine-mapped, providing key biological insights. Most studies have been conducted in populations of European ancestry, but many studies now include multi-ethnic samples. Analysis of multi-ethnic data presents many advantages, including increased power to detect associated variants that are rare or absent in Europeans and increased resolution for fine-mapping, but also many challenges. The extent to which genetic architectures are shared across ethnicities is not well-understood, the implications for meta-analyzing studies across ethnicities are uncertain, and the optimal strategy for performing fine-mapping in multi-ethnic data remains an open question, particularly when allowing for multiple causal variants at a locus. These challenges can inhibit multi-ethnic study designs, limiting opportunities to detect new associations and address health disparities in minority populations. Here, we propose to develop a complete set of methods and software for disease mapping in multi-ethnic populations, building on the extensive progress of our research program over the past four years. Our goal is to make fully powered association and fine-mapping studies as practical in multi-ethnic populations as in studies of a single continental population. Our methods research will be driven by empirical data from >900,000 samples (>700,000 with raw genotypes/phenotypes and >200,000 with summary statistics), including African American, Latino, East Asian and South Asian samples spanning a wide range of quantitative and disease phenotypes. We will develop methods for both raw genotype/phenotype data and summary association statistic data, and the methods will be applicable to both common and rare variation, including gene-based tests.